The curves in the graph were taken in our anechoic chamber with no smoothing applied. We do apply a correction below 80 Hz to take into account chamber inaccuracies that occur below this frequency but we do not do the final design of the low frequencies in the anechoic chamber. These are done on the tower. After the bass design is complete utilizing the tower we can write a custom correction for the chamber to give us an exact replica of the tower measurements below 80 Hz. This allows us to get the whole picture from the measurements taken in the anechoic chamber.

Designing a speaker is all about the family of curves. There are around 70 basic curves to consider when doing a speaker design and then, as you start to factor in various microphone positions, this number grows in multiples and you can easily end up with over 200 curves to consider. Understanding the family of curves and their relationship to the listening experience is the key to designing great-sounding speakers. This is where Double Blind Listen Testing is so invaluable. The only way to know how changes to various curves affect the end listening experience is the empirical test results that come from endless Double Blind Listen Testing.

The omnidirectional multi-channel design of the LFR1100 gives us much more control over the entire family of curves: a control which is made possible by having the DSP in the mix. The LFR1100 is not designed to be 100% omnidirectional which would mean the sound power and the listening window would be essentially the same curve, and not one you would want to have equal energy at all frequencies. It is instead a blend of front-firing only and 100% omnidirectional designs, a sort of cherry picking of the best attributes of both designs.

Ian Colquhoun
President & Chief Engineer